I'm trying to update our application from Compass 2.0.0M1 to 2.0.0M2. I'm holding all other things constant and only changing the Compass and Lucene jars. I'm recreating the search index for our data and I'm seeing deadlock in Lucene's IndexWriter. It appears to be waiting on a single from the merge thread. (This is called from Compass's TransIndex.commit().) When I look at the other Compass threads that are running (Compass Scheduled Executor and two Compass Executor threads), none of them are active, they're waiting on work, and none of them indicate they might be a merge thread. When I look at the sub index directory, there's a write.lock there, though it's not clear which thread is holding the lock. I'm deleting the entire index directory before running this scenario to ensure there's no existing lock. Doing the exact same steps with 2.0.0M1 has no problems and recreates the search index. I haven't had a chance to dig deeper into Compass / Lucene on this. Is it a known problem?